Generative Adversarial Networks and Image Synthesis

Últimos artículos

• The Ensemble Inverse Problem: Applications and Methods

  Zhengyan Huan, Camila Pazos, Martin Klassen, Vincent Croft, Pierre-Hugues Beauchemin, Shuchin Aeron · 30 de enero de 2026

  We introduce a new multivariate statistical problem that we refer to as the Ensemble Inverse Problem (EIP). The aim of EIP is to invert for an ensemble that is distributed according to the pushforward of a prior under a forward process. In high energy physics (HEP), this is related to a widely known…

• Generation Enhances Understanding in Unified Multimodal Models via Multi-Representation Generation

  Zihan Su, Hongyang Wei, Kangrui Cen, Yong Wang, Guanhua Chen, Chun Yuan, Xiangxiang Chu · 30 de enero de 2026

  Unified Multimodal Models (UMMs) integrate both visual understanding and generation within a single framework. Their ultimate aspiration is to create a cycle where understanding and generation mutually reinforce each other. While recent post-training methods have successfully leveraged understanding…

• Unsupervised Decomposition and Recombination with Discriminator-Driven Diffusion Models

  Archer Wang, Emile Anand, Yilun Du, Marin Solja\v{c}i\'c · 30 de enero de 2026

  Decomposing complex data into factorized representations can reveal reusable components and enable synthesizing new samples via component recombination. We investigate this in the context of diffusion-based models that learn factorized latent spaces without factor-level supervision. In images, facto…

• Visual Disentangled Diffusion Autoencoders: Scalable Counterfactual Generation for Foundation Models

  Sidney Bender, Marco Morik · 30 de enero de 2026

  Foundation models, despite their robust zero-shot capabilities, remain vulnerable to spurious correlations and 'Clever Hans' strategies. Existing mitigation methods often rely on unavailable group labels or computationally expensive gradient-based adversarial optimization. To address these limitatio…

• A Diffusive Classification Loss for Learning Energy-based Generative Models

  Louis Grenioux, RuiKang OuYang, Jos\'e Miguel Hern\'andez-Lobato · 30 de enero de 2026

  Score-based generative models have recently achieved remarkable success. While they are usually parameterized by the score, an alternative way is to use a series of time-dependent energy-based models (EBMs), where the score is obtained from the negative input-gradient of the energy. Crucially, EBMs …

• Gauge-invariant representation holonomy

  Vasileios Sevetlidis, George Pavlidis · 30 de enero de 2026

  Deep networks learn internal representations whose geometry--how features bend, rotate, and evolve--affects both generalization and robustness. Existing similarity measures such as CKA or SVCCA capture pointwise overlap between activation sets, but miss how representations change along input paths. …

• The Depth Delusion: Why Transformers Should Be Wider, Not Deeper

  Md Muhtasim Munif Fahim, Md Rezaul Karim · 30 de enero de 2026

  Neural scaling laws describe how language model loss decreases with parameters and data, but treat architecture as interchangeable--a billion parameters could arise from a shallow-wide model (10 layers & 8,192 hidden dimension) or a deep-narrow one (80 layers & 2,048 hidden dimension). We propose ar…

• Rethinking Refinement: Correcting Generative Bias without Noise Injection

  Xin Peng, Ang Gao · 30 de enero de 2026

  Generative models, including diffusion and flow-based models, often exhibit systematic biases that degrade sample quality, particularly in high-dimensional settings. We revisit refinement methods and show that effective bias correction can be achieved as a post-hoc procedure, without noise injection…

• Signal from Structure: Exploiting Submodular Upper Bounds in Generative Flow Networks

  Alexandre Larouche, Audrey Durand · 30 de enero de 2026

  Generative Flow Networks (GFlowNets; GFNs) are a class of generative models that learn to sample compositional objects proportionally to their a priori unknown value, their reward. We focus on the case where the reward has a specified, actionable structure, namely that it is submodular. We show subm…

• FlexCausal: Flexible Causal Disentanglement via Structural Flow Priors and Manifold-Aware Interventions

  Yutao Jin, Yuang Tao, Junyong Zhai · 30 de enero de 2026

  Causal Disentangled Representation Learning(CDRL) aims to learn and disentangle low dimensional representations and their underlying causal structure from observations. However, existing disentanglement methods rely on a standard mean-field approximation with a diagonal posterior covariance, which d…

• Generative Modeling through Koopman Spectral Analysis: An Operator-Theoretic Perspective

  Yuanchao Xu, Fengyi Li, Masahiro Fujisawa, Xiaoyuan Cheng, Youssef Marzouk, Isao Ishikawa · 30 de enero de 2026

  We propose Koopman Spectral Wasserstein Gradient Descent (KSWGD), a particle-based generative modeling framework that learns the Langevin generator via Koopman theory and integrates it with Wasserstein gradient descent. Our key insight is that this spectral structure of the underlying distribution c…

• Memorization Control in Diffusion Models from Denoising-centric Perspective

  Thuy Phuong Vu, Mai Viet Hoang Do, Minhhuy Le, Dinh-Cuong Hoang, Phan Xuan Tan · 30 de enero de 2026

  Controlling memorization in diffusion models is critical for applications that require generated data to closely match the training distribution. Existing approaches mainly focus on data centric or model centric modifications, treating the diffusion model as an isolated predictor. In this paper, we …

• Bi-Anchor Interpolation Solver for Accelerating Generative Modeling

  Hongxu Chen, Hongxiang Li, Zhen Wang, Long Chen · 30 de enero de 2026

  Flow Matching (FM) models have emerged as a leading paradigm for high-fidelity synthesis. However, their reliance on iterative Ordinary Differential Equation (ODE) solving creates a significant latency bottleneck. Existing solutions face a dichotomy: training-free solvers suffer from significant per…

• Entropy-Based Dimension-Free Convergence and Loss-Adaptive Schedules for Diffusion Models

  Ahmad Aghapour, Erhan Bayraktar, Ziqing Zhang · 30 de enero de 2026

  Diffusion generative models synthesize samples by discretizing reverse-time dynamics driven by a learned score (or denoiser). Existing convergence analyses of diffusion models typically scale at least linearly with the ambient dimension, and sharper rates often depend on intrinsic-dimension assumpti…

• From Logits to Latents: Contrastive Representation Shaping for LLM Unlearning

  Haoran Tang, Rajiv Khanna · 30 de enero de 2026

  Most LLM unlearning methods aim to approximate retrain-from-scratch behaviors with minimal distribution shift, often via alignment-style objectives defined in the prediction space. While effective at reducing forgotten content generation, such approaches may act as suppression: forgotten concepts ca…

• Flow Perturbation++: Multi-Step Unbiased Jacobian Estimation for High-Dimensional Boltzmann Sampling

  Xin Peng, Ang Gao · 30 de enero de 2026

  The scalability of continuous normalizing flows (CNFs) for unbiased Boltzmann sampling remains limited in high-dimensional systems due to the cost of Jacobian-determinant evaluation, which requires $D$ backpropagation passes through the flow layers. Existing stochastic Jacobian estimators such as th…

• Multilevel and Sequential Monte Carlo for Training-Free Diffusion Guidance

  Aidan Gleich, Scott C. Schmidler · 30 de enero de 2026

  We address the problem of accurate, training-free guidance for conditional generation in trained diffusion models. Existing methods typically rely on point-estimates to approximate the posterior score, often resulting in biased approximations that fail to capture multimodality inherent to the revers…

• Investigating Associational Biases in Inter-Model Communication of Large Generative Models

  Fethiye Irmak Dogan, Yuval Weiss, Kajal Patel, Jiaee Cheong, Hatice Gunes · 30 de enero de 2026

  Social bias in generative AI can manifest not only as performance disparities but also as associational bias, whereby models learn and reproduce stereotypical associations between concepts and demographic groups, even in the absence of explicit demographic information (e.g., associating doctors with…

• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks

  Luwei Sun, Dongrui Shen, Jianfe Li, Yulong Zhao, Han Feng · 30 de enero de 2026

  Motivated by challenges in conditional generative modeling, where the target conditional density takes the form of a ratio f1 over f2, this paper develops a theoretical framework for approximating such ratio-type functionals. Here, f1 and f2 are kernel-based marginal densities that capture structure…

• Mitigating data replication in text-to-audio generative diffusion models through anti-memorization guidance

  Francisco Messina, Francesca Ronchini, Luca Comanducci, Paolo Bestagini, Fabio Antonacci · 30 de enero de 2026

  A persistent challenge in generative audio models is data replication, where the model unintentionally generates parts of its training data during inference. In this work, we address this issue in text-to-audio diffusion models by exploring the use of anti-memorization strategies. We adopt Anti-Memo…

• Optimization and Mobile Deployment for Anthropocene Neural Style Transfer

  Po-Hsun Chen, Ivan C. H. Liu · 30 de enero de 2026

  This paper presents AnthropoCam, a mobile-based neural style transfer (NST) system optimized for the visual synthesis of Anthropocene environments. Unlike conventional artistic NST, which prioritizes painterly abstraction, stylizing human-altered landscapes demands a careful balance between amplifyi…

• Revisiting Diffusion Model Predictions Through Dimensionality

  Qing Jin, Chaoyang Wang · 30 de enero de 2026

  Recent advances in diffusion and flow matching models have highlighted a shift in the preferred prediction target -- moving from noise ($\varepsilon$) and velocity (v) to direct data (x) prediction -- particularly in high-dimensional settings. However, a formal explanation of why the optimal target …

• SimGraph: A Unified Framework for Scene Graph-Based Image Generation and Editing

  Thanh-Nhan Vo, Trong-Thuan Nguyen, Tam V. Nguyen, Minh-Triet Tran · 30 de enero de 2026

  Recent advancements in Generative Artificial Intelligence (GenAI) have significantly enhanced the capabilities of both image generation and editing. However, current approaches often treat these tasks separately, leading to inefficiencies and challenges in maintaining spatial consistency and semanti…

• DreamActor-M2: Universal Character Image Animation via Spatiotemporal In-Context Learning

  Mingshuang Luo, Shuang Liang, Zhengkun Rong, Yuxuan Luo, Tianshu Hu, Ruibing Hou, Hong Chang, Yong Li, Yuan Zhang, Mingyuan Gao · 30 de enero de 2026

  Character image animation aims to synthesize high-fidelity videos by transferring motion from a driving sequence to a static reference image. Despite recent advancements, existing methods suffer from two fundamental challenges: (1) suboptimal motion injection strategies that lead to a trade-off betw…

• Generative Modeling of Discrete Data Using Geometric Latent Subspaces

  Daniel Gonzalez-Alvarado, Jonas Cassel, Stefania Petra, Christoph Schn\"orr · 30 de enero de 2026

  We introduce the use of latent subspaces in the exponential parameter space of product manifolds of categorial distributions, as a tool for learning generative models of discrete data. The low-dimensional latent space encodes statistical dependencies and removes redundant degrees of freedom among th…